• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Amd fx8350 multiplier changing throttling back cpu

Associate
Joined
11 Mar 2014
Posts
285
Hi guys, I've been redoing my overclock on my cpu for 2 reasons, I have double the cooling capacity I did have and I noticed during stability tests be it prime95 or amd overdrive after about 10m the multiplier goes from where I've set it (19.5) to 7 reducing the cpu speed and it bounces back and forth between the two. Sometimes 1 core sometimes several. It does it at 4.6, 4.7 and 4.8ghz where previously it was 100% stable. The temps have not been above 66 deg c socket and 50 deg c core @4.8ghz and no more than 60 socket and 41 core @4.6ghz so I'm pretty sure it's not a thermal problem.

My bios is set as per rog site cool n quiet disabled, c1e disabled etc. power settings all extreme and 130%. I read the multiplier changing thread which although intel based suggested sometimes a particular fsb/multi combo can cause instability so I've tried the same over clocks using a multi of 19.5, 20 and 23 for 4.6ghz same result each time.

Any suggestions for a fix for this please as I said above I had it running previously @4.8ghz no problems.

Many thanks in advance, Damien.
 
In what way have you doubled cooling capacity?

When I had AMD FX I found that going from air to water created its own problems in that the motherboard VRM's weren't getting any active cooling, maybe your new cooler is not cooling the VRM's as good?
 
I had 2 x120mm rads front and rear with no cable management. I've moved to a fractal midi r2 with proper cable management and 2 x 240 rads. I can hold my finger on the nb/Vrm heatsink until you get bored and I've got a temp probe on it that's never been above 35 deg c.

Originally before the custom cooling I had a corsair hydro h60 in push pull that covered the Vrm heatsink whereas now with just a rear mounted fan the Vrm is open to airflow. Same when I mounted the 120 in the rear of my old case. Just 1 fan so the Vrm heatsink was open.

If that is the problem I'm assuming I can slacken off the heatsink slide the probe in then tighten it down again so I can measure the temp as hwmonitor just says a constant 128 deg c. Apparently it's a common fault.
 
Maybe you need a touch more volts in your current configuration. Are you using offset or specified vcore? When I had my M5A99x v1, I used LLC one step below the max and 130% CPU and 120% CPU/NB power settings. What HT frequency are you using?
 
I think I may have found the problem I had cool n quiet set to 'disabled by cpu' which was what the online guide I followed specified. Changed to always disabled and although still early days of testing it seems ok at the mo. I'm going to post all my settings and a link to a google doc of my test results when I'm done in the fx8350 thread.
 
Scratch that I hadn't found the problem but I have now. The Vrm heatsink overheating. I pointed a 120mm sickleflow at it and it stopped throttling back. Whilst running the amd overdrive stability test I then took the fan away and about a minute later it throttled. Reintroduced the fan and hey presto clocks went back up again. You would've thought the 3rd best asus board with pro in the title could hack 4.6ghz or higher wouldn't you.
 
Scratch that I hadn't found the problem but I have now. The Vrm heatsink overheating. I pointed a 120mm sickleflow at it and it stopped throttling back. Whilst running the amd overdrive stability test I then took the fan away and about a minute later it throttled. Reintroduced the fan and hey presto clocks went back up again. You would've thought the 3rd best asus board with pro in the title could hack 4.6ghz or higher wouldn't you.

nothing to do with the board, its down to your cooling, ppl put water is a pc and say yep that's done now it can get any better..... but you need air movement over so many components and water cooling removes the flow of air

Here is a nice selection of fans to help you fix the problem, i have 1 40mm on NB and 2x40mm on Vrm
http://www.overclockers.co.uk/productlist.php?groupid=701&catid=2331&subid=818
 
It's not really Asus' fault more AMD's for releasing ridiculously power hungry processors, that's one of the reason I prefer Intel you don't have to worry about VRM cooling and throttling.
 
I got around this issue on my AMD test board by using RAM fan coolers and bending the clips to fit the sinks.

My CHVF-Z bards sinks get quite toasty though if i put on the original aluminium sinks, not enough to throttle but too toasty for my comfort.

I love watercooling so much now, that my last four rigs have had VRM blocks mainly for the aesthetics.
 
Last edited:
It's not really Asus' fault more AMD's for releasing ridiculously power hungry processors, that's one of the reason I prefer Intel you don't have to worry about VRM cooling and throttling.

Im sorry but you make out to be worse than it is. My old sabertooth board had no issues supplying power to a 5.0 ghz FX-8350 and the vrms never got over 79 degrees and that was just with the stock heatsink on them and a proper fan setup in my case like every normal person would do.
 
Im sorry but you make out to be worse than it is. My old sabertooth board had no issues supplying power to a 5.0 ghz FX-8350 and the vrms never got over 79 degrees and that was just with the stock heatsink on them and a proper fan setup in my case like every normal person would do.

Yes the problem is down to water cooling and not thinking about cooling other parts
 
Im sorry but you make out to be worse than it is. My old sabertooth board had no issues supplying power to a 5.0 ghz FX-8350 and the vrms never got over 79 degrees and that was just with the stock heatsink on them and a proper fan setup in my case like every normal person would do.

I would say 2x120 sp120 performance in the front, 2x120 sp120 performance and 1 x fractal 140 in the roof and 1 x sp120 quiet in the rear would be classed as a normal fan setup. Fractal and fronts intake, 120's in roof and rear exhausting. Psu upside down drawing hot air from the case too.

Have full cable management so the air has nothing blocking it from the front to the rear. I'm going to try these little fans over the heatsink failing that ill bung this motherboard and ram into a system I'm building to sell and get a crosshair v formula and get a waterblock for the board as well.

It runs for approx 7m of any stability test be it prime, ibt or amd overdrive and the multiplier goes from whatever it's set to down 7 and back up then after a min stays at 7. I can't exaggerate what's actually happening. If it wasn't that bad I wouldn't be asking for help.
 
Back
Top Bottom